Introduction to Information Retrieval INFS 550 WebBased Information

Introduction to Information Retrieval INFS 550 Web-Based Information Retrieval System Chapter 1 Introduction to Information Retrieval

Introduction to Information Retrieval § Information Retrieval (IR) is finding material (usually documents) of an unstructured nature (usually text) that satisfies an information need from within large collections (usually stored on computers). 2

Introduction to Information Retrieval Definition § Information Retrieval (IR) is the task of finding relevant texts within a large amount of unstructured data Ø Relevant ≡ texts matching some specific criteria. Ø Examples of IR tasks: searching for emails from a given person, searching for an event that occured on a given date using the Internet, etc. Ø Examples of IR systems: www search engines, specific search engines (laws, medical documents), etc. Ø NB: Data. Bases Management Systems (DBMS) are different from IR systems (data stored in a DB are structured!) 3

Introduction to Information Retrieval An IR task is a 3 -step process: 1. Asking a question (how to use the language to get what we want ? ) 2. Building an answer from known data (how to refer to a given text ? ) 3. Assessing the answer (does it contain the information we were seeking ? ) Remarks: Ø This model can loop, the assessing phase then allows to improve the quality of the answers (more on this later). Ø The user may not be able to define exactly the question (i. e. characterize his/her information needs). 4

Introduction to Information Retrieval Aspects that have to be taken into account from the IR system’s point of view: Ø How to allow to ask questions about information we are not aware of ? Ø What if the question is ill-formed (aborting, looking for document matching close questions, asking for precisions) ? Ø What if there is no answer (automatic modification of the question) ? Ø What is the best format for the answer (single answer, ordered list of answers, etc) ? Ø How to be sure that the answer will be understood (vocabulary used, etc) ? All these questions have to be taken into account in the realization of a search engine. 5

Introduction to Information Retrieval 6

Introduction to Information Retrieval Some terminology Ø An IR system looks for data matching some criteria defined by the users in their queries. Ø The language used to ask a question is called the query language. Ø These queries use keywords (atomic items characterizing some data). Ø The basic unit of data is a document (can be a file, an article, a paragraph, etc. ). Ø A document corresponds to free text (may be unstructured). Ø All the documents are gathered into a collection (or corpus). Example: 1 million documents, each counting about 1000 words if each word is encoded using 6 bytes: 109 × 1000 × 6/1024 ≃ 6 GB 7

Introduction to Information Retrieval Sec. 1. 1 Unstructured data in 1680 § Which plays of Shakespeare contain the words Brutus AND Caesar but NOT Calpurnia? § One could grep all of Shakespeare’s plays for Brutus and Caesar, then strip out lines containing Calpurnia? § Why is that not the answer? § Slow (for large corpora) § NOT Calpurnia is non-trivial § Other operations (e. g. , find the word Romans near countrymen) not feasible § Ranked retrieval (best documents to return) § Later lectures 8

Introduction to Information Retrieval Indexing documents § how to relate the user’s information need with some documents’ content ? § idea: using an index to refer to documents § Usually an index is a list of terms that appear in a document, it can be represented mathematically as: § Here, the kind of index we use maps keywords to the list of documents they appear in: § we call this an inverted index. 9

Introduction to Information Retrieval Indexing documents ØThe set of keywords is usually called the dictionary (or vocabulary). ØA document identifier appearing in the list associated with a keyword is called a posting. ØThe list of document identifiers associated with a given keyword is called a posting list. 10

Introduction to Information Retrieval Indexing documents Ø Arising questions: how to build an index automatically ? What are the relevant keywords ? (more on this later) Ø Some additional objectives: • fast processing of large collections of documents, • having flexible matching operations (robust retrieval), • having the possibility to rank the retrieved document in terms of relevance. Ø to ensure these requirements (especially fast processing) are fulfilled, the indexes are computed in advance. Note that the format of the index has a huge impact on the performances of the system. 11

Introduction to Information Retrieval Indexing documents NB: an index is built in 4 steps: 1. Gathering of the collection (each document is given a unique identifier). 2. Segmentation of each document into a list of atomic tokens (referred to as the tokenization operation). 3. Linguistic processing of the tokens in order to normalize them. 4. Indexing the documents by computing the dictionary and lists of postings 12

Sec. 1. 1 Introduction to Information Retrieval Term-document incidence Brutus AND Caesar BUT NOT Calpurnia 1 if play contains word, 0 otherwise

Introduction to Information Retrieval Sec. 1. 1 Incidence vectors § So we have a 0/1 vector for each term. § To answer query: take the vectors for Brutus, Caesar and Calpurnia (complemented) bitwise AND. § 110100 AND 110111 AND 101111 = 100100. 14

Introduction to Information Retrieval Sec. 1. 1 Answers to query § Antony and Cleopatra, Act III, Scene ii Agrippa [Aside to DOMITIUS ENOBARBUS]: Why, Enobarbus, When Antony found Julius Caesar dead, He cried almost to roaring; and he wept When at Philippi he found Brutus slain. § Hamlet, Act III, Scene ii Lord Polonius: I did enact Julius Caesar I was killed i' the Capitol; Brutus killed me. 15

Introduction to Information Retrieval Sec. 1. 1 Basic assumptions of Information Retrieval § Collection: Fixed set of documents § Goal: Retrieve documents with information that is relevant to the user’s information need and helps the user complete a task 16

Introduction to Information Retrieval The classic search model Get rid of mice in a politically correct way TASK Misconception? Info about removing mice without killing them Info Need Mistranslation? Verbal form How do I trap mice alive? Misformulation? Query mouse trap SEARCH ENGINE Query Refinement Results Corpus

Introduction to Information Retrieval Sec. 1. 1 How good are the retrieved docs? § Precision : Fraction of retrieved docs that are relevant to user’s information need § Recall : Fraction of relevant docs in collection that are retrieved § More precise definitions and measurements to follow in later lectures 18

Introduction to Information Retrieval Sec. 1. 1 Bigger collections § Consider N = 1 million documents, each with about 1000 words. § Avg 6 bytes/word including spaces/punctuation § 6 GB of data in the documents. § Say there are M = 500 K distinct terms among these. 19

Introduction to Information Retrieval Sec. 1. 1 Can’t build the matrix § 500 K x 1 M matrix has half-a-trillion 0’s and 1’s. Why? § But it has no more than one billion 1’s. § matrix is extremely sparse. § What’s a better representation? § We only record the 1 positions. 20

Sec. 1. 2 Introduction to Information Retrieval Inverted index § For each term t, we must store a list of all documents that contain t. § Identify each by a doc. ID, a document serial number § Can we use fixed-size arrays for this? Brutus 1 Caesar 1 Calpurnia 2 2 2 31 4 11 31 45 173 174 4 5 6 16 57 132 54 101 What happens if the word Caesar is added to document 14? 21

Sec. 1. 2 Introduction to Information Retrieval Inverted index § We need variable-size postings lists § On disk, a continuous run of postings is normal and best § In memory, can use linked lists or variable length arrays § Some tradeoffs in size/ease of insertion Brutus 1 Caesar 1 Calpurnia Dictionary 2 2 2 31 Posting 4 11 31 45 173 174 4 5 6 16 57 132 54 101 Postings Sorted by doc. ID (more later on why). 22

Sec. 1. 2 Introduction to Information Retrieval Inverted index construction Documents to be indexed Friends, Romans, countrymen. Tokenizer Token stream More on these later. Modified tokens Inverted index Friends Romans Linguistic modules friend roman Countrymen countryman Indexer friend 2 4 roman 1 2 countryman 13 16

Sec. 1. 2 Introduction to Information Retrieval Indexer steps: Token sequence § Sequence of (Modified token, Document ID) pairs. Doc 1 I did enact Julius Caesar I was killed i' the Capitol; Brutus killed me. Doc 2 So let it be with Caesar. The noble Brutus hath told you Caesar was ambitious

Introduction to Information Retrieval Indexer steps: Sort § Sort by terms § And then doc. ID Core indexing step Sec. 1. 2

Introduction to Information Retrieval Sec. 1. 2 Indexer steps: Dictionary & Postings § Multiple term entries in a single document are merged. § Split into Dictionary and Postings § Doc. frequency information is added. Why frequency? Will discuss later.

Sec. 1. 2 Introduction to Information Retrieval Where do we pay in storage? Lists of doc. IDs Terms and counts Pointers Later in the course: • How do we index efficiently? • How much storage do we need? 27

Introduction to Information Retrieval Sec. 1. 3 The index we just built § How do we process a query? § Later - what kinds of queries can we process? Today’s focus 28

Sec. 1. 3 Introduction to Information Retrieval Query processing: AND § Consider processing the query: Brutus AND Caesar § Locate Brutus in the Dictionary; § Retrieve its postings. § Locate Caesar in the Dictionary; § Retrieve its postings. § “Merge” the two postings: 2 4 8 16 1 2 3 5 32 8 64 13 128 21 Brutus 34 Caesar 29

Sec. 1. 3 Introduction to Information Retrieval The merge § Walk through the two postings simultaneously, in time linear in the total number of postings entries 2 8 2 4 8 16 1 2 3 5 32 8 128 64 13 21 Brutus 34 Caesar If list lengths are x and y, merge takes O(x+y) operations. Crucial: postings sorted by doc. ID. 30

Introduction to Information Retrieval Intersecting two postings lists (a “merge” algorithm) 31

Introduction to Information Retrieval Sec. 1. 3 Boolean queries: Exact match § The Boolean retrieval model is being able to ask a query that is a Boolean expression: § Boolean Queries use AND, OR and NOT to join query terms § Views each document as a set of words § Is precise: document matches condition or not. § Perhaps the simplest model to build an IR system on § Primary commercial retrieval tool for 3 decades. § Many search systems you still use are Boolean: § Email, library catalog, Mac OS X Spotlight 32

Sec. 1. 4 Introduction to Information Retrieval Example: West. Law http: //www. westlaw. com/ § Largest commercial (paying subscribers) legal search service (started 1975; ranking added 1992) § Tens of terabytes of data; 700, 000 users § Majority of users still use boolean queries § Example query: § What is the statute of limitations in cases involving the federal tort claims act? § LIMIT! /3 STATUTE ACTION /S FEDERAL /2 TORT /3 CLAIM § /3 = within 3 words, /S = in same sentence 33

Sec. 1. 4 Introduction to Information Retrieval Example: West. Law http: //www. westlaw. com/ § Another example query: § Requirements for disabled people to be able to access a workplace § disabl! /p access! /s work-site work-place (employment /3 place) § Note that SPACE is disjunction, not conjunction! § Long, precise queries; proximity operators; incrementally developed; not like web search § Many professional searchers still like Boolean search § You know exactly what you are getting § But that doesn’t mean it actually works better….

Introduction to Information Retrieval Boolean queries: More general merges Sec. 1. 3 § Exercise: Adapt the merge for the queries: Brutus AND NOT Caesar Brutus OR NOT Caesar Can we still run through the merge in time O(x+y)? What can we achieve? 35

Introduction to Information Retrieval Sec. 1. 3 Merging What about an arbitrary Boolean formula? (Brutus OR Caesar) AND NOT (Antony OR Cleopatra) § Can we always merge in “linear” time? § Linear in what? § Can we do better? 36

Sec. 1. 3 Introduction to Information Retrieval Query optimization § What is the best order for query processing? § Consider a query that is an AND of n terms. § For each of the n terms, get its postings, then AND them together. Brutus 2 Caesar 1 Calpurnia 4 2 8 16 32 64 128 3 5 8 16 21 34 13 16 Query: Brutus AND Calpurnia AND Caesar 37

Sec. 1. 3 Introduction to Information Retrieval Query optimization example § Process in order of increasing freq: § start with smallest set, then keep cutting further. This is why we kept document freq. in dictionary Brutus 2 Caesar 1 Calpurnia 4 2 8 16 32 64 128 3 5 8 16 21 34 13 16 Execute the query as (Calpurnia AND Brutus) AND Caesar. 38

Introduction to Information Retrieval Sec. 1. 3 More general optimization § e. g. , (madding OR crowd) AND (ignoble OR strife) § Get doc. freq. ’s for all terms. § Estimate the size of each OR by the sum of its doc. freq. ’s (conservative). § Process in increasing order of OR sizes. 39

Introduction to Information Retrieval Exercise § Recommend a query processing order for (tangerine OR trees) AND (marmalade OR skies) AND (kaleidoscope OR eyes) 40

Introduction to Information Retrieval Query processing exercises § Exercise: If the query is friends AND romans AND (NOT countrymen), how could we use the freq of countrymen? § Exercise: Extend the merge to an arbitrary Boolean query. Can we always guarantee execution in time linear in the total postings size? § Hint: Begin with the case of a Boolean formula query where each term appears only once in the query. 41

Introduction to Information Retrieval Exercise § Try the search feature at http: //www. rhymezone. com/shakespeare/ § Write down five search features you think it could do better 42

Introduction to Information Retrieval What’s ahead in IR? Beyond term search § What about phrases? § Stanford University § Proximity: Find Gates NEAR Microsoft. § Need index to capture position information in docs. § Zones in documents: Find documents with (author = Ullman) AND (text contains automata). 43

Introduction to Information Retrieval Evidence accumulation § 1 vs. 0 occurrence of a search term § 2 vs. 1 occurrence § 3 vs. 2 occurrences, etc. § Usually more seems better § Need term frequency information in docs 44

Introduction to Information Retrieval Ranking search results § Boolean queries give inclusion or exclusion of docs. § Often we want to rank/group results § Need to measure proximity from query to each doc. § Need to decide whether docs presented to user are singletons, or a group of docs covering various aspects of the query. 45

Introduction to Information Retrieval IR vs. databases: Structured vs unstructured data § Structured data tends to refer to information in “tables” Employee Manager Salary Smith Jones 50000 Chang Smith 60000 Ivy Smith 50000 Typically allows numerical range and exact match (for text) queries, e. g. , Salary < 60000 AND Manager = Smith. 46

Introduction to Information Retrieval Unstructured data § Typically refers to free-form text § Allows § Keyword queries including operators § More sophisticated “concept” queries, e. g. , § find all web pages dealing with drug abuse § Classic model for searching text documents 47

Introduction to Information Retrieval Semi-structured data § In fact almost no data is “unstructured” § E. g. , this slide has distinctly identified zones such as the Title and Bullets § Facilitates “semi-structured” search such as § Title contains data AND Bullets contain search … to say nothing of linguistic structure 48

Introduction to Information Retrieval More sophisticated semi-structured search § Title is about Object Oriented Programming AND Author something like stro*rup § where * is the wild-card operator § Issues: § how do you process “about”? § how do you rank results? § The focus of XML search (IIR chapter 10) 49

Introduction to Information Retrieval Clustering, classification and ranking § Clustering: Given a set of docs, group them into clusters based on their contents. § Classification: Given a set of topics, plus a new doc D, decide which topic(s) D belongs to. § Ranking: Can we learn how to best order a set of documents, e. g. , a set of search results 50

Introduction to Information Retrieval The web and its challenges § Unusual and diverse documents § Unusual and diverse users, queries, information needs § Beyond terms, exploit ideas from social networks § link analysis, clickstreams. . . § How do search engines work? And how can we make them better? 51

Introduction to Information Retrieval More sophisticated information retrieval § § § Cross-language information retrieval Question answering Summarization Text mining … 52

Introduction to Information Retrieval Course details § Course URL: cs 276. stanford. edu § [a. k. a. , http: //www. stanford. edu/class/cs 276/ ] § Work/Grading: § § Problem sets (2) Practical exercises (2) Midterm Final 20% 10% + 20% = 30% 20% 30% § Textbook: § Introduction to Information Retrieval § In bookstore and online (http: //informationretrieval. org/) § We’re happy to get comments/corrections/feedback on it! 53

Introduction to Information Retrieval Course staff § Professor: Pandu Nayak nayak@cs. stanford. edu § Professor: Prabhakar Raghavan pragh@cs. stanford. edu § TAs: Sonali Aggarwal, Sandeep Sripada, Valentin Spitkovsky § In general, don’t use the above addresses, but: § Newsgroup: su. class. cs 276 [preferred] § cs 276 -spr 1011 -staff@lists. stanford. edu 54

Introduction to Information Retrieval Resources for today’s lecture § Introduction to Information Retrieval, chapter 1 § Shakespeare: § http: //www. rhymezone. com/shakespeare/ § Try the neat browse by keyword sequence feature! § Managing Gigabytes, chapter 3. 2 § Modern Information Retrieval, chapter 8. 2 Any questions? 55